chemistry - American Chemical Society

parasites (Worthing and Hance, 1991). Its utilization is undoubtedly beneficial, but there are toxicological and environmental risks associated with t...
0 downloads 0 Views 509KB Size
JUNE 1994 VOLUME 42, NUMBER 6

Iournal of Agricultural andW

chemistry

0 Copyright 1994 by the American Chemical Society

SHORT COMMUNICATIONS Development of a Chlorpyrifos Immunoassay Using Antibodies Obtained from a Simple Hapten Design INTRODUCTION Chlorpyrifos [O,O-diethyl0-(3,5,6-trichloro-2-pyridinyl) phosphorothioatel is a broad spectrum organophosphorus insecticide, widely used for the control of various crop pests in soil and on foliage, household insects, and animal parasites (Worthing and Hance, 1991). Its utilization is undoubtedly beneficial, but there are toxicological and environmental risks associated with the remaining residues after its application. Increasing chlorpyrifos and other organophosphate pesticide occurrence, along with recent concerns over long-term/low-level exposure to agrochemicals, has led to more comprehensive environmental and food monitoring programs (Vanderlaan et al., 1990;Yess et al., 1993). Chlorpyrifos analyses are carried out by chromatographic techniques, mainly gas chromatography (Iwata et al., 1983;Sanz et al., 1991). Chromatographic methods are laborious and time-consuming and require sophisticated equipment only available in well-equipped centralized laboratories. Immunoassay technology is being demonstrated as a rapid and cost-effective alternative to traditional methods when a high sample throughput or on-site screening analyses is required (Vanderlaan et al., 1990;Sherry, 1992). For the production of antibodies to small molecules such as chlorpyrifos, suitable haptens have to be synthesized and conjugated to carrier proteins, to render the analyte structure immunogenic. In aromatic organophosphate pesticides, two basic moieties can be differentiated for conjugation purposes: a phosphate group and an aromatic ring. An elaborated generic strategy for the synthesis of organothiophosphate haptens, with spacer arm attachment through the thiophosphate moiety, has been recently published (McAdam et al., 1992). Using this hapten, an immunoassay for the analysis of chlorpyrifos-methyl in cereal matrices, with appropriate sensitivity and specificity, was developed (Skerritt et al., 1992). On the other hand, organophosphate hapten immunogens prepared by modification through the aromatic ring rendered antibod0021-058 If9411442-1257$04.50/ 0

ies with relatively poor recognition properties of the target analyte (Vallejoet al., 1982;Brimfield e t al., 1985;McAdam et al., 1992). In the present study, we describe the one-step synthesis of a chlorpyrifos hapten, which consists of the attachment of a spacer arm through the aromatic ring with minor modification of the chemical properties of the original structure. From protein conjugates of this hapten, several monoclonal antibodies (MAb) that recognize chlorpyrifos were obtained. We also report the use of these MAb in the development of a sensitive and specific competitive enzyme immunoassay for the analysis of chlorpyrifos. EXPERIMENTAL PROCEDURES Materials and Instruments. Organic synthesis reagents, standards, and biological products were of the highest grade availablefrom current commercialsuppliers. 'H and S ' C nuclear magnetic resonance (NMR) spectra were recorded with a Varian VXR-400s spectrometer (Varian, Sunnyvale, CA), operating at 400 and 100MHz, respectively. Absorbances obtained in enzymelinked immunosorbent assays (ELISA) were read and recorded with a Dynatech MR-700 microplate reader (Sussex, U.K.). Hapten Synthesis. The structures of haptens 1 and 2 are shown in Figure 1. Synthetic intermediates were analyzed by TLC and confirmed by lH NMR. 0,O-Diethyl 0-[3,5-Dichloro-6-[(2-carboryethyZ)thio]-2pyridyl] Phosphorothioate (I). To a solution of 3-mercaptopropanoic acid (1.06 g, 10 mmol) in 50 mL of absolute ethanol, 2 equiv of KOH (1.42 g) was added and heated until dissolved. Then, chlorpyrifos (technical grade, 3.51 g, 10 mmol) dissolved in 50 mL of absolute ethanol was added. After reflux for 1h, the reaction mixture was filtered and the solvent was removed under reduced pressure. To the residue was added 5% NaHCOs (50 mL) followed by washing with hexane (2 X 50 mL). The aqueous layer was acidified to pH 4 and extracted with dichloromethane (3 x 50 mL). The extract was dried over NazSO, and concentrated, and the residue was subjected to column chromatography [hexane/tetrahydrofuran (THF)/aceticacid 7525:ll. Fractions showing only one spot on TLC (Rf0.41, same solvent mixture) were pooled and concentratedto provide 1.07g of 1 (26%), which 0 1994 American Chemical Society

1258 J. Agric. Food Chem., Vol. 42, No. 6, 1994

Short

Communications

precipitation followed by anion-exchange chromatography on DEAE-Sepharose (Sigma, St. Louis, MO). El-P-0 Enzyme-Linked Immunosorbent Assay. Flat bottom C hlorpyrifos I BO polystyrene ELISA plates (Nunc, Roskilde, Denmark) were coated in carbonate buffer overnight at 4 OC. Antibodies were diluted in phosphate-buffered saline containing 0.05% Tween 20 (PBST). A volume of 100 pL per well was used throughout SCH,CH,COOH all assay steps. After each step, plates were washed with PBST. Indirect ELISA. Plates coated with the OVA-1 conjugate (1 BO-P-0 Hapten 1 pg/mL) were incubated with culture supernatants or serial El0 CI dilutionsof sera at room temperaturefor 1h. Captured antibodies were measured by incubation at room temperature for 1h with goat anti-mouse peroxidase-labeled immunoglobulins, followed NHCH,CH,COOH by the addition of the substrate solution (2 mg/mL o-phenyleneI diamine, 0.012 % HzO2in phosphate-citrate buffer, pH 5.4).After 10 min, the enzymatic reaction was stopped by adding H2SO4 and the absorbance was read at 490 nm. BO-P-0I Hapten 2 Competitiue Indirect ELISA. The protocol was the same as Et0 for the indirect ELISA except that after coating with OVA-1 or OVA-2 (homologous or heterologous assay, respectively), a Figure 1. Structures of chlorpyrifos and the haptens used for competition step was established by adding the appropriate preparing conjugates. concentration of antibody (serum,culture supernatant,or purified MAb) to different concentrations of competitor (chlorpyrifosor solidified on standing: mp 124-125 "C; lH NMR (CDCls)6 7.76 related compounds). For each assay, the optimal concentrations (S, lH, ArH), 4.37 (q+q, 4H, 2 CH20), 3.45 (t,2H, SCH2), 2.97(t, of both coating conjugate and antibody were estimated by 2H, CHaCOO), 1.42 (t, 6H, 2 CH3); 13C NMR (CDCl3) 6 177.1 checkerboard titration. Competition curves were obtained by (COOH),153.8 (C6,Ar), 151.6 (C2, Ar), 139.2 (C4,Ar), 125.2 $5, Ar),116.2(C3,Ar),65.4(2OCH2),34.2(SCH2),25.4(CH2COOH),plotting absorbance vs the logarithm of competitor concentration and were fitted to four-parameter logistic equations (Raab,1983). 15.9 (2 CH3). 0,O-Diethyl 0-[3,4,6Trichloro-2-[[(2-carboxyethyl)amiRESULTS AND DISCUSSION no]carbonyl]phenyl] Phosphorothioate (2). The protected amino acid tert-butyl 3-aminopropanoate was synthesized as The chlorpyrifos structure (Figure 1)contains a t least described by McAdam et al. (1992).To a stirred solution of 3,5,6two sites for spacer attachment. One option is the trichlorosalicylicacid (486 mg, 2 mmol), tert-butyl3-aminoprointroduction of the spacer as a thiophosphate ester. This panoate (580mg,4 mmol),and (dimethy1amino)pyridine(20mg) approach was successfullyapplied by McAdam et al. (1992) in 3 mL of THF was added dicyclohexylcarbodiimide (DCC;453 mg, 2.2 mmol) in 1mL of THF. After 5 h at room temperature, for the development of a chlorpyrifos-methyl immunoasthe mixture wasfiltered and the solvent evaporatedunder reduced say, but a multistep synthesis procedure to prepare the pressure. The yellowish oil was subjected to column chromathiophosphoramide haptens was followed. The second tography (hexane/ethyl acetate/acetic acid 70:301) to provide possibility, which is studied here, is spacer coupling 246 mg (0.667 mmol) of tert-butyl 3-[(2-hydroxy-3,5,6-trichlothrough the pyridyl ring by chlorine substitution using a ropheny1)methanamidolpropanoate (33% ,Rj0.63) as a colorless suitable nucleophilic compound. In this sense, 3-mersolid. To a solution of this product in THF/water (l:l,5 mL) captopropanoic acid was previously used in a similar case was added 1M NaOH (667 pL). The solvent was removed under for synthesizing s-triazine herbicide haptens (Goodrow et reduced pressure and the residue was heated at 80 "C until al., 1990). complete dryness to give the corresponding sodium salt. To a heterogeneous mixture of this salt in 8 mL of acetonitrile was Hapten 1 (Figure 1)was directly prepared from chloadded diethyl chlorothiophosphate (150 mg, 0.8 mmol). The rpyrifos by reaction with 3-mercaptopropanoic acid. Its mixture was refluxed for 1h and then filtered, and the solvent structure was confirmed by lH and 13CNMR. Special was evaporated. Without further purification, the residue was analytical attention was directed to finding out which treated with 50% trifluoroaceticacid in 2 mL of dichloromethane chlorine (ring position 3, 5, or 6) was preferentially for 1h to deprotect the carboxylic group. After evaporation of substituted. Position 6 is activated for nucleophilic the solvent, the residue was chromatographed (hexane/ethyl substitution by the nitrogen in ortho. In fact, attempts acetate/acetic acid 50:501) to yield 157 mg of 2 (51%) (Rj0.54 to carry out the same reaction with the chlorpyrifos analog in TLC, same solvent mixture): mp 161-162OC; lH NMR (CDCl3) fenchlorphos-ethyl, containing carbon instead of nitrogen 6 7.57 (S, lH, ArH), 6.71 (S, lH, NH), 4.27 (q+q, 4H, 2 CH20), as the unique structural modification, failed to provide 3.69 (d, 2H, NCHz), 2.75 (t, 2H, CH&OO), 1.38 (t, 6H, 2 CH3). Preparation of Hapten-Protein Conjugates. Haptens 1 detectable quantities of any substituted compound. This and 2 were coupled to proteins (bovine serum albumin, BSA; result confirmed the strong activation exerted by the ovalbumin, OVA) by the active ester method described by nitrogen on the adjacent chlorine. Furthermore, by Langone and Van Vunakis (1975). Conjugates (BSA-1, OVA-1, comparing the 13C NMR chemical shifts of chlorpyrifos and OVA-2) were purified by gel filtration chromatography on and hapten 1, a clear displacement of the signal assigned Sephadex G-25 (Pharmacia, Uppsala, Sweden). to C6 (McAdam and Skerritt, 1993) from 144.2 to 153.8 Production of Monoclonal Antibodies. Five BALB/c ppm was observed. Therefore, spacer attachment was very female mice (8-10 weeks old) were intraperitoneally immunized likely accomplished in the chlorine in position 6. with 2 week-interval injections of the immunogen (BSA-1, 100 Using the active ester method, suitable hapten-protein pg per mouse)emulsifiedwith Freund's adjuvants. Mouse serum titers and analyte recognition properties were analyzed during conjugates were obtained. A clear spectrum modification the immunization scheduleby indirect ELISA. Splenocytesfrom of protein-hapten 1 conjugates a t 315 nm after coupling mice showing the highest serum titers were fused with P3-X63/ allowed reliable estimates of the hapten to protein molar Ag8.653 murine myeloma cells (ATCC,Rockville, MA) according ratio (23 and 8 for BSA-1 and OVA-1, respectively). to established protocols (Nowinski et al., 1979). Hybridoma Five mice were immunized with BSA-1. After three culture supernatants were screened by indirect ELISA for injections, serum titers (serum dilution that gave 3 times antibodies binding to OVA-1. Positive hybridomas were further background absorbance) in the range 1:100OoO-1:3OO000 analyzed by competitive indirect ELISA, and those selected were were found by indirect ELISA, using the homologous cloned by limiting dilutions. MAb were purified from late OVA-1 as coating conjugate. Serum titers did not stationary phase culture supernatants by ammonium sulfate

$=-&

ba

J. Agrlc. FoodChem., Vol. 42, No. 6, 1994

Short Communications

Table 1. Summary of the Monoclonal Antibody Selection Results no. of wells no. of cloned seeded poaitiveb competitivec hybridomas 1 384 18 8 4 2 384 48 3 1 “Mice used for fusion 1 or 2 received three or five biweekly intraperitoneal injections of BSA-1, respectively. Wells with antibodies that recognized the OVA-1 coating conjugate by indirect ELISA (absorbance >0.5). Wells with antibodies that recognized free-chlorpyrifos (inhibition >50% by 3 or 1 pM chlorpyrifos, for fusion 1 or 2, respectively). Competitive ELISAs were carried out with 0.3 pg/mL of OVA-1 as coating and the appropriatesupernatant dilutions giving absorbances in the range 0.8-1.5.

1258

Table 2. Cross-Reactivity of Related Compounds in the Heterologous Competitive Indirect ELISA for Chlorpyrifos

fusion no.4 ~~

1 . 0 p /

,

,

,

,

,

,

,

,

compound chlorpyrifos chlorpyrifos-methyl fenchlorphos bromophos trichloronate dichlofenthion parathion 3,5,6-trichloro2-pyridinol (metabolite)

Ri

Et0 Me0 Me0 Et0 Et0 Et0 Et0

structure x Rs Et0 N.C1 Me0 N C1 Me0 C C1 Et0 C C1 Et C C1 Et0 C H Et0 C H

Rz

& C1 C1 C1 Br C1 C1 NO2

ICw, CR,4 nM % C1 33 100 C1 18 183 C1 835 3.8 C1 1760 1.8 C1 2880 1.1 C1 4950 0.67 H 20700 0.16 2160 1.5 R.5

4 Percentage of cross-reactivity= (ICwof chlorpyrifrm/IC~ of other compound) X 100.

0.4

o.2 0.0

L o

1 0 - * i o - ~ ioo

io1

A

i o e l o 3 10‘

io6

[CHLORPYRIFOS] (nM)

Figure 2. ELISA standard curves of chlorpyrifos using LIB-C4 homologous assay (0.2 pg/mL of OVA-1 as coating, MAb: (0) 0.05 rg/mL of MAb); ( 0 )heterologous assay (1 pg/mL of OVA-2 as coating, 0.2 pg/mL of MAb). Each point represents the mean f SD for eight replicates.

significantly improve after five injections. Sera were subsequently tested for their ability to bind chlorpyrifos by indirect competitive ELISA. The concentration producing 50% inhibition of antibody binding (1Cw) was 2.5 pM chlorpyrifos, confirming the usefulness of the designed hapten to elicit antibodies that recognize the free analyte. Two fusions were carried out using splenocytes from the mice showing the highest serum titers. The results of the hybridoma culture supernatant screening are presented in Table 1. Of a total of 768 wells seeded, 66 were considered positive by indirect ELISA. Of the 6 6 , l l were selected by a restrictive indirect competitive ELISA (inhibition >50% with 3 pM chlorpyrifos). Finally, five hybridoma cell lines were cloned and stabilized. MAb were purified and characterized for chlorpyrifos recognition properties. One of them, LIB-C4, showed the highest affinity for chlorpyrifos (IC50 = 197 nM, 69 ng/mL) in a homologous indirect competitive ELISA (Figure 2). This result, obtained from a simple hapten derivatized through the aromatic ring, is comparable to those previously reported for chlorpyrifos-methyl haptens derivatized through the thiophosphate group (Skerritt et al., 1992). Heterologous assays are well-known procedures to improve hapten immunoassay sensitivity (Harrison et al., 1991). Attachment position and spacer modification have been proved to be adequate types of heterology (Wie and Hammock, 1984; Szurdoki et al., 1992). The heterologous hapten2 (Figure 11,with both site and spacer heterologies, was synthesized as described under Experimental Procedures and coupled to OVA with an approximate molar ratio of 12. The OVA-2 conjugate was subsequently used as coating antigen in a heterologous indirect competitive

ELISA. The typical standard curve of chlorpyrifos using L I B 4 4 in the heterologous assay is shown in Figure 2. With this ELISA format, sensitivity was improved 6 times (IC50 = 33 nM, 11.6 ng/mL for chlorpyrifos). The heterologous assay was further characterized regarding its limit of detection (LOD) and specificity. The assay LOD, calculated as the chlorpyrifos concentration corresponding to the absorbance of the zero dose minus three standard deviations, was 3.7 nM (1.3 ng/mL). Specificity was evaluated by performing competitive assays with several related organophosphorus insecticides as competitors. ICWand relative cross-reactivity (CR) values for each compound are given in Table 2. Chlorpyrifosmethyl showed an even lower ICs0 (18 nM, 183% CR), suggesting that the methyl groups are better accommodated in the antibody binding site. Another important feature was that the change from pyridyl to phenyl-type compounds markedly influences the interaction, since other halogenated organothiophosphates bearing this change cross-reacted less than 2% (bromophos, 1.8%; trichloronate, 1.1% ;dichlofenthion, 0.67% 1, except fenchlorphos (3.8%) which is in consonance with being a methyl analog. CR of other tested organophosphateswere